INFECOMP : Inferential processes development : On-line investigations during reading comprehension
This proposal concerns basic school learning, in particular reading comprehension. His originality is based on the angle adopted to study the development of the skills necessary for comprehension in a reading situation : he proposes to take a systematic interest not in the product of the activity of understanding (ie, the mental representation) but to the cognitive processes that operate throughout this activity. Specifically, it focuses on the study of the development of a precise subset of processes : the processes underlying the production of inferences during reading, all researchers agreeing on the idea that inference production is an essential skill for comprehension in reading situations.
While current research sheds light on the development of the ability to produce inferences, current methods are not relevant to examine the process itself. Faced with this pitfall, this project proposes to free itself from the commonly used methodology, which essentially consists in evaluating the understanding of texts once comprehension had been entirely completed. Rather than evaluating the inferences that the child has generated at the end of his reading, it is a question of privileging a real-time approach allowing the study of the cognitive processes at work in the production of inferences at the time they operate.
To fulfill our goal, we embed our thinking and assumptions into the RI-VAL model recently proposed by Cook and O’Brien (2015, O’Brien & Cook, 2016a, 2016b). The main hypothesis tested in this project is then that differences in the accuracy and speed of access to information (factor 1) combined with differences in the coherence threshold (factor 2) are two factors that contribute to the development of the ability to produce inferences. Our research program aims to determine the extent to which these factors condition the production of inferences during its development at each of the phases underlying the production of an inference, as defined in the RI- VAL (ie, activation, integration, validation).
We propose to study the development of inferential processes in reading comprehension as children proceed through the text, using three experimental methodologies conventionally used in psychology to identify causal links :
- an analysis of the general development of inferential processes via two 2 years cross-sectional longitudinal studies within two cohorts : 5-7 and 8-10 years old children ;
- a differential approach of interindividual trajectories, based on a comparison between less-skilled and skilled comprehenders ;
- Finally, in order to prevent or remedy the difficulties of understanding, we will develop and test the effectiveness of an intervention program on inferences production, the approach of which would be differentiated according to comprehender profiles (Bianco et al., 2012, Ecalle et al., 2015, Potocki et al., 2015).
To summarize, the present proposal aims at investigating cognitive processes that underlie inference production while children comprehend a text in order to provide new information on how this ability develops. By examining on-line cognitive processes, it will contribute to improve our knowledge on how children acquired reading comprehension abilities but it will also enable a deeper and more reliable understanding of difficulties that children encounter when learning to comprehend a text. The results will point toward concrete directions to investigate in order to avoid that initial differences in efficiency develop into reading comprehension difficulties fastidious to overcome.
La coordinatrice principale du projet est Sabine Géraud, Prof. en psychologie cognitive à UP8 (DYSCO).
Le projet a débuté en décembre 2017 pour une durée de 48 mois.